<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	>
<channel>
	<title>Comments on: What have we achieved so far?</title>
	<atom:link href="http://www.gnucitizen.org/blog/what-have-we-achieved-so-far/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.gnucitizen.org/blog/what-have-we-achieved-so-far/</link>
	<description>Information Security Think Tank</description>
	<pubDate>Mon, 06 Oct 2008 13:09:57 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.6.1</generator>
		<item>
		<title>By: Ross Snider</title>
		<link>http://www.gnucitizen.org/blog/what-have-we-achieved-so-far/#comment-122958</link>
		<dc:creator>Ross Snider</dc:creator>
		<pubDate>Mon, 14 Jul 2008 12:54:14 +0000</pubDate>
		<guid isPermaLink="false">http://www.gnucitizen.org/?p=895#comment-122958</guid>
		<description>@sal-e

I don't think anyone else is worried, and you should now be either. I'll solve your insomnia right now.

An RSA key of 1024 bits has to be factored to be broken. We rely on the difficulty of this task in order to keep our transmissions secure.

It takes an estimated 10^12 MIPS*years to break an RSA key of this size. At a terraflop a piece (1,000,000 MIPS) it would take 1 computer around a million years to break RSA. Now, if we assume our culprit can buy 1,000,000 such computers and wire/program them to work together, it could be cracked in one year.

Thankfully this is unlikely to happen. No worm is going to be spread that can take advantage of the full CPU and also not get noticed and taken out. The budget of the individual behind this would in the hundred millions or billions at least.

So lets say we really get scared of this happening. We'll up standard key size. Against a 2048 bit key, this scenario looks (2^1024 times) even more unplausible. Against a 4096 bit key, this looks absurd.

Before any one can break RSA, we'll have moved onto elliptic curve.</description>
		<content:encoded><![CDATA[<p>@sal-e</p>
<p>I don&#8217;t think anyone else is worried, and you should now be either. I&#8217;ll solve your insomnia right now.</p>
<p>An RSA key of 1024 bits has to be factored to be broken. We rely on the difficulty of this task in order to keep our transmissions secure.</p>
<p>It takes an estimated 10^12 MIPS*years to break an RSA key of this size. At a terraflop a piece (1,000,000 MIPS) it would take 1 computer around a million years to break RSA. Now, if we assume our culprit can buy 1,000,000 such computers and wire/program them to work together, it could be cracked in one year.</p>
<p>Thankfully this is unlikely to happen. No worm is going to be spread that can take advantage of the full CPU and also not get noticed and taken out. The budget of the individual behind this would in the hundred millions or billions at least.</p>
<p>So lets say we really get scared of this happening. We&#8217;ll up standard key size. Against a 2048 bit key, this scenario looks (2^1024 times) even more unplausible. Against a 4096 bit key, this looks absurd.</p>
<p>Before any one can break RSA, we&#8217;ll have moved onto elliptic curve.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: pdp</title>
		<link>http://www.gnucitizen.org/blog/what-have-we-achieved-so-far/#comment-122943</link>
		<dc:creator>pdp</dc:creator>
		<pubDate>Sun, 13 Jul 2008 05:05:21 +0000</pubDate>
		<guid isPermaLink="false">http://www.gnucitizen.org/?p=895#comment-122943</guid>
		<description>Greg, try getting a course out of your field of work. Seriously. You might find it more enjoyable.</description>
		<content:encoded><![CDATA[<p>Greg, try getting a course out of your field of work. Seriously. You might find it more enjoyable.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: sal-e</title>
		<link>http://www.gnucitizen.org/blog/what-have-we-achieved-so-far/#comment-122937</link>
		<dc:creator>sal-e</dc:creator>
		<pubDate>Sat, 12 Jul 2008 18:47:15 +0000</pubDate>
		<guid isPermaLink="false">http://www.gnucitizen.org/?p=895#comment-122937</guid>
		<description>Hi PDP,

There is a lot of noise around Web2.0 security. But I am observing fundamental shift of the personal computer hardware that will make the current security practices obsolete virtually over night. At the same time I don't hear anything about it.

Any person that keeps track of the latest video hardware knows that Nvidia just released a GPU with 1.4 billion transistors. Its performance is estimated to be about 1 Teraflops. That is a super computer. The main purpose of the GPU is to decode video and graphical information. At the same time there is war between Intel and GPU manufactures like AMD and Nvidia. As result of this war the GPU manufactures are opening their hardware to be used for general computing tasks. There are prototypes from Adobe for video transcoding accelerated by the GPU and results are very impressive. Those applications are still not available, but there is a project called "Folding@Home" (http://folding.stanford.edu). They already have high performance client that are using GPU. The GPUs are many, many time faster then regular CPUs in some tasks. What is the common between video and folding proteins is that both are very scalable across parallel GPU. As far as my limited knowledge there is one more task that shares the same characteristics - encoding and decoding any data, especially in the case of missing encryption keys. Every one knows that any encryption can be broken if you have unlimited time and computation power. Now what will happen when someone writes an encryption cracking tool that uses the latest Nvidia GPU?!  

Is anybody else having sleepless night because of it?</description>
		<content:encoded><![CDATA[<p>Hi PDP,</p>
<p>There is a lot of noise around Web2.0 security. But I am observing fundamental shift of the personal computer hardware that will make the current security practices obsolete virtually over night. At the same time I don&#8217;t hear anything about it.</p>
<p>Any person that keeps track of the latest video hardware knows that Nvidia just released a GPU with 1.4 billion transistors. Its performance is estimated to be about 1 Teraflops. That is a super computer. The main purpose of the GPU is to decode video and graphical information. At the same time there is war between Intel and GPU manufactures like AMD and Nvidia. As result of this war the GPU manufactures are opening their hardware to be used for general computing tasks. There are prototypes from Adobe for video transcoding accelerated by the GPU and results are very impressive. Those applications are still not available, but there is a project called &#8220;Folding@Home&#8221; (http://folding.stanford.edu). They already have high performance client that are using GPU. The GPUs are many, many time faster then regular CPUs in some tasks. What is the common between video and folding proteins is that both are very scalable across parallel GPU. As far as my limited knowledge there is one more task that shares the same characteristics - encoding and decoding any data, especially in the case of missing encryption keys. Every one knows that any encryption can be broken if you have unlimited time and computation power. Now what will happen when someone writes an encryption cracking tool that uses the latest Nvidia GPU?!  </p>
<p>Is anybody else having sleepless night because of it?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Greg</title>
		<link>http://www.gnucitizen.org/blog/what-have-we-achieved-so-far/#comment-122934</link>
		<dc:creator>Greg</dc:creator>
		<pubDate>Sat, 12 Jul 2008 13:57:30 +0000</pubDate>
		<guid isPermaLink="false">http://www.gnucitizen.org/?p=895#comment-122934</guid>
		<description>Hey pdp, how about your passing the TIGER exam.  I though you guys were the only TIGER team out there, or am I wrong?
Out of interest did you and ap find it tough?  I'm trying to choose a course to do but they seem way to tough for me right now :(.  I'm not a security jock, just some guy in IT who wants to break stuff.</description>
		<content:encoded><![CDATA[<p>Hey pdp, how about your passing the TIGER exam.  I though you guys were the only TIGER team out there, or am I wrong?<br />
Out of interest did you and ap find it tough?  I&#8217;m trying to choose a course to do but they seem way to tough for me right now :(.  I&#8217;m not a security jock, just some guy in IT who wants to break stuff.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: LKP</title>
		<link>http://www.gnucitizen.org/blog/what-have-we-achieved-so-far/#comment-122930</link>
		<dc:creator>LKP</dc:creator>
		<pubDate>Sat, 12 Jul 2008 10:04:43 +0000</pubDate>
		<guid isPermaLink="false">http://www.gnucitizen.org/?p=895#comment-122930</guid>
		<description>Well, for what matters, a lot you guys have accomplished, there will be always more to be done, is just the way of things, not only in computer related stuff but in life aswell.

There are a lot of people who in some point just look back to see they didn't accomplished nothing, so, no matter if it was one thing or two, I know you'll feel better when you look at your past to see all the things you accomplished as a group.

It's even better that beside all the things you've done you still have that motivation of keep doing new things so keep the work! ;)</description>
		<content:encoded><![CDATA[<p>Well, for what matters, a lot you guys have accomplished, there will be always more to be done, is just the way of things, not only in computer related stuff but in life aswell.</p>
<p>There are a lot of people who in some point just look back to see they didn&#8217;t accomplished nothing, so, no matter if it was one thing or two, I know you&#8217;ll feel better when you look at your past to see all the things you accomplished as a group.</p>
<p>It&#8217;s even better that beside all the things you&#8217;ve done you still have that motivation of keep doing new things so keep the work! ;)</p>
]]></content:encoded>
	</item>
</channel>
</rss>
